VIIth International Scientific and Technical Conference From Imagery to Map: Digital Photogrammetric Technologies Master class: Master class: Photogrammetric processing of Photogrammetric processing of pushbroom satellite images pushbroom satellite images Petr S. Titarov, Software Developer September 17-20, 2007, Nessebar, Bulgaria
43
Embed
VIIth International Scientific and Technical Conference From Imagery to Map: Digital Photogrammetric Technologies Master class: Photogrammetric processing.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
VIIth International Scientific and Technical Conference From Imagery to Map: Digital Photogrammetric Technologies
Master class:Master class:Photogrammetric processing of Photogrammetric processing of pushbroom satellite imagespushbroom satellite images
I. Basics of space pushbroom imagingI. Basics of space pushbroom imaging
Photogrammetric processing problemsPhotogrammetric processing problems Methods of pushbroom photogrammetryMethods of pushbroom photogrammetry
II. Pushbroom photogrammetryII. Pushbroom photogrammetry
Systems of resolution 1 meter and betterSystems of resolution 1 meter and better Systems of resolution about 2 metersSystems of resolution about 2 meters Systems of resolution 5 metersSystems of resolution 5 meters Systems of resolution 10-20 metersSystems of resolution 10-20 meters
III. Pushbroom imaging systems overviewIII. Pushbroom imaging systems overview
V. Satellite pushbroom imagery processing using PHOTOMODV. Satellite pushbroom imagery processing using PHOTOMOD
IV. Taking choice of remote sensing product for photogrammetryIV. Taking choice of remote sensing product for photogrammetry
Geometry of pushbroom imagery significantly differs from central projection one,Geometry of pushbroom imagery significantly differs from central projection one,so the classical photogrammetric methods are not applicableso the classical photogrammetric methods are not applicable..
Problems solved to perform photogrammetric processingProblems solved to perform photogrammetric processing Methods of pushbroom photogrammetryMethods of pushbroom photogrammetry
Pushbroom photogrammetryPushbroom photogrammetry
),,(
),,(),,(
),,(
4
3
2
1
NNN
NNNN
NNN
NNNN
hP
hPy
hP
hPx
4,1,3
0
3
0
3
0),,(
3
qi j k
kN
jN
iNijkqhP ha
kji
NNNq
r2
r1
S1
S2
Orbit
Earth
G
S1
S2
M
ray 1ray 2
Space resection and space intersectionSpace resection and space intersection
GCPs
Single images
DEM3D vectors
Orthoimagery2D vectors
DEM
GCPs
Stereopairs
Ortho-mosaics
Export to GIS, CAD,
digital maps
right
right
left
left
y
x
y
x
f
Z
Y
X
Z
Y
X
gy
x
Methods of pushbroom photogrammetryMethods of pushbroom photogrammetry
Rigorous Replacement models
Methods of pushbroom photogrammetry
Modeling imagery acquisition geometry
Using abstract relationships that approximate rigorous
imaging model
Generic
Using a-priory relationships containing parameters determined from GCPs
The model defines the directional vector of ray sensed by the detector number p with respect to the sensor reference system S:
The model is the analog of interior orientation elements in classical photogrammetry.
Sensor motion modelSensor motion model
Polynomial modelPolynomial model Orbital modelOrbital model
Z
Y
X
n
k
kk
n
j
jj
n
i
ii
lCZ
lBY
lAX
0
0
0
Keplerian orbit parameters:• major semi-axis a • eccentricity e• inclination i• celestial longitude of the ascending node • argument of perigee • perigee passing time
Parameters to be refined:
e, i, ,, sometimes a
Parameters to be refined:
Ai , Bj , Ck
applicable in any Cartesian reference system (including ECR) simple to implement
inertial reference system must be used physical model only a few parameters to refine
Sensor attitude modelSensor attitude model
The model is defined by the three angles , , , which polynomially depend on the line number l, or it can be represented as the sum of the measured in-flight values and polynomial refinements:
K
k
kkèçì
J
j
jjèçì
I
i
iièçì
lll
lll
lll
0
0
0
)()(
)()(
)()(
The model defines the rotation of the sensorreference system with respect to the geocentricReference system.
))(),(),(()( lllPlP
Rigorous solution of space resection and space Rigorous solution of space resection and space intersection intersection
Space intersectionSpace intersection Space resectionSpace resection
Z
Y
X
fy
x
2
2
1
1
y
x
y
x
g
Z
Y
X
Correspondent rays intersection Iterative process
Imagery orientationImagery orientation
0)( SMrSMr
Z
Y
X
SM
Z
Y
X
Pr
,
0
0
0
XYYX
ZXXZ
YZZY
Any two of the three equations are independent:
The image orientation is based on the collinearity condition:
Using some a-priory equations derived from coarse assumptions concerning imaging geometry whichrelate image coordinates x, y to ground ones X,Y,Z.The values of the parameters involved into the equations are calculated using GCPs.
),...,,,,(
),...,,,,(
1
1
m
n
qqZYXgy
ppZYXfx
Direct Linear Transformation (DLT)
1
1
321
4321
321
4321
ZCYCXC
BZBYBXBy
ZCYCXC
AZAYAXAx
Parallel-perspective model
1
1
321
321
4321
ZCYCXCy
ZBYBXB
AZAYAXAx
Replacement models
They are models which approximateground-to-image correspondencecalculated using rigorous method:
TIFF raster andTIFF raster and RPCRPC Super Structured file format, Super Structured file format, processing using generic methodsprocessing using generic methods
PartPart VV
Satellite pushbroom imagery processing using PHOTOMODSatellite pushbroom imagery processing using PHOTOMOD